Book Image

Clojure Data Analysis Cookbook - Second Edition

By : Eric Richard Rochester
Book Image

Clojure Data Analysis Cookbook - Second Edition

By: Eric Richard Rochester

Overview of this book

Related Content you might be interested in

Current Title:

Small book image
Clojure Data Analysis Cookbook - Second Edition
Eric Richard Rochester
Jan, 2015 | 372 pages
No titles found
Table of Contents (19 chapters)
Clojure Data Analysis Cookbook Second Edition
Clojure Data Analysis Cookbook Second Edition
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Importing Data for Analysis
Importing Data for Analysis
Introduction
Creating a new project
Reading CSV data into Incanter datasets
Reading JSON data into Incanter datasets
Reading data from Excel with Incanter
Reading data from JDBC databases
Reading XML data into Incanter datasets
Scraping data from tables in web pages
Scraping textual data from web pages
Reading RDF data
Querying RDF data with SPARQL
Aggregating data from different formats
2
Cleaning and Validating Data
Cleaning and Validating Data
Introduction
Cleaning data with regular expressions
Maintaining consistency with synonym maps
Identifying and removing duplicate data
Regularizing numbers
Calculating relative values
Parsing dates and times
Lazily processing very large data sets
Sampling from very large data sets
Fixing spelling errors
Parsing custom data formats
Validating data with Valip
3
Managing Complexity with Concurrent Programming
Managing Complexity with Concurrent Programming
Introduction
Managing program complexity with STM
Managing program complexity with agents
Getting better performance with commute
Combining agents and STM
Maintaining consistency with ensure
Introducing safe side effects into the STM
Maintaining data consistency with validators
Monitoring processing with watchers
Debugging concurrent programs with watchers
Recovering from errors in agents
Managing large inputs with sized queues
4
Improving Performance with Parallel Programming
Improving Performance with Parallel Programming
Introduction
Parallelizing processing with pmap
Parallelizing processing with Incanter
Partitioning Monte Carlo simulations for better pmap performance
Finding the optimal partition size with simulated annealing
Combining function calls with reducers
Parallelizing with reducers
Generating online summary statistics for data streams with reducers
Using type hints
Benchmarking with Criterium
5
Distributed Data Processing with Cascalog
Distributed Data Processing with Cascalog
Introduction
Initializing Cascalog and Hadoop for distributed processing
Querying data with Cascalog
Distributing data with Apache HDFS
Parsing CSV files with Cascalog
Executing complex queries with Cascalog
Aggregating data with Cascalog
Defining new Cascalog operators
Composing Cascalog queries
Transforming data with Cascalog
6
Working with Incanter Datasets
Working with Incanter Datasets
Introduction
Loading Incanter's sample datasets
Loading Clojure data structures into datasets
Viewing datasets interactively with view
Converting datasets to matrices
Using infix formulas in Incanter
Selecting columns with $
Selecting rows with $
Filtering datasets with $where
Grouping data with $group-by
Saving datasets to CSV and JSON
Projecting from multiple datasets with $join
7
Statistical Data Analysis with Incanter
Statistical Data Analysis with Incanter
Introduction
Generating summary statistics with $rollup
Working with changes in values
Scaling variables to simplify variable relationships
Working with time series data with Incanter Zoo
Smoothing variables to decrease variation
Validating sample statistics with bootstrapping
Modeling linear relationships
Modeling non-linear relationships
Modeling multinomial Bayesian distributions
Finding data errors with Benford's law
8
Working with Mathematica and R
Working with Mathematica and R
Introduction
Setting up Mathematica to talk to Clojuratica for Mac OS X and Linux
Setting up Mathematica to talk to Clojuratica for Windows
Calling Mathematica functions from Clojuratica
Sending matrixes to Mathematica from Clojuratica
Evaluating Mathematica scripts from Clojuratica
Creating functions from Mathematica
Setting up R to talk to Clojure
Calling R functions from Clojure
Passing vectors into R
Evaluating R files from Clojure
Plotting in R from Clojure
9
Clustering, Classifying, and Working with Weka
Clustering, Classifying, and Working with Weka
Introduction
Loading CSV and ARFF files into Weka
Filtering, renaming, and deleting columns in Weka datasets
Discovering groups of data using K-Means clustering
Finding hierarchical clusters in Weka
Clustering with SOMs in Incanter
Classifying data with decision trees
Classifying data with the Naive Bayesian classifier
Classifying data with support vector machines
Finding associations in data with the Apriori algorithm
10
Working with Unstructured and Textual Data
Working with Unstructured and Textual Data
Introduction
Tokenizing text
Finding sentences
Focusing on content words with stoplists
Getting document frequencies
Scaling document frequencies by document size
Scaling document frequencies with TF-IDF
Finding people, places, and things with Named Entity Recognition
Mapping documents to a sparse vector space representation
Performing topic modeling with MALLET
Performing naïve Bayesian classification with MALLET
11
Graphing in Incanter
Graphing in Incanter
Introduction
Creating scatter plots with Incanter
Graphing non-numeric data in bar charts
Creating histograms with Incanter
Creating function plots with Incanter
Adding equations to Incanter charts
Adding lines to scatter charts
Customizing charts with JFreeChart
Customizing chart colors and styles
Saving Incanter graphs to PNG
Using PCA to graph multi-dimensional data
Creating dynamic charts with Incanter
12
Creating Charts for the Web
Creating Charts for the Web
Introduction
Serving data with Ring and Compojure
Creating HTML with Hiccup
Setting up to use ClojureScript
Creating scatter plots with NVD3
Creating bar charts with NVD3
Creating histograms with NVD3
Creating time series charts with D3
Visualizing graphs with force-directed layouts
Creating interactive visualizations with D3
Index
Index

Classifying data with support vector machines

Support vector machines (SVMs) try to divide two groups of data along a plane. An SVM finds the plane that is the farthest from both groups. If a plane comes much closer to group B, it will prefer a plane that is approximately an equal distance from both. SVMs have a number of nice properties. While other clustering or classification algorithms work well with defined clusters of data, SVMs may work fine with data that isn't in well-defined and delineated groupings. They are also not affected by the local minima. Algorithms such as K-Means or SOMs—which begin from a random starting point—can get caught in solutions that aren't bad for the area around the solution, but aren't the best for the entire space. This isn't a problem for SVMs.

Getting ready

First, we'll need these dependencies in our project.clj file:

(defproject d-mining "0.1.0-SNAPSHOT"
  :dependencies [[org.clojure/clojure "1.6.0"]
                 [nz.ac.waikato.cms.weka/weka-dev "3...
Previous Section
End of Section 10
Next Section